Thursday, December 27, 2007

Seven Flags Deconstructed

Thanks for the comments to yesterday’s quiz. You all explained flag behavior better than I could. The seven flags are all variations on the “wavy ribbon” idea, which is how we think flags should look.

In fairness, there’s nothing wrong with this way of drawing flags—if your goal is to represent the mental image of a flag. A cartoonist is usually after the mental image, not reality. It’s also a good way to represent a flag if you want to show the flag's graphic design clearly.

But if your goal is realism, it’s worth observing that flags never actually appear with undulating folds parallel to the flagpole. Instead, as many of you pointed out, a set of folds radiates diagonally downward from the upper point of support.

I was unaware of this principle until a day in 1995 when I was stuck in rest stop during a long bus trip through the midlands of England. There was nothing to do but sketch and nothing to sketch but a flag. I drew a page of variations as the wind changed from a zephyr to a stiff breeze.

Here are some YouTube videos, which show waving flags better than my sketches.
Medium size flag in diminishing wind: Link
Small flag in stronger wind: Link
Small flag in heavy wind Link
Big flag in heavy wind: Link ; (In this last one, the diagonal rule breaks down a bit, and the wrinkles are more complex billows. They actually start to look a bit like the flag pictures from China and France, above. I have a feeling the math behind all this is pretty complex.)

Perhaps one of our CG animation friends might be willing to say a few words about the challenges involved in modeling this kind of action in 3D on a computer.

Well, it's hard to pick. The grand prize has to go to the first Anonymous ( who got the basic answer right away. The next four runners up who really described the action are: Kevin H, Dan G., Orlando M., and Meredith D. But I'm going to give the last runner up prize to Big E, who expressed the larger truth that there's no single correct way to represent reality; it depends what you want to communicate. If you’d like to collect your prizes, please email your mailing address to me at


Anonymous said...

This has been an interesting and discussion and rewarding in itself. No need to send me a prize.

I've enjoyed about your discussions about the effort to achieve realism in art. This effort requires the practitioner to observe more carefully and process those observations much more mindfully. I think this is one way in which art helps us to literally discover the miraculous world in which we live. In this way, art can be a similar to Buddhist mindfulness practice. Achieving a more accurate, mindful connection to the world is, for me, a noble reason to pick up a pencil or paintbrush.

On the other hand, "painting what you think you see," as a cartoonist might do, offers a window into how the mind conceptualizes and simplifies the natural world. Studying the art of young children can be interesting in this regard.

Thinking about how and why this simplification occurs and what it means interests me. Perhaps you can think about it and post your thoughts someday.

The important thing for all of us to do (are you listening, George B.?) is to notice the difference between our conceptualizations and reality itself.

Anonymous said...

Yes, very interesting... Allow me point out ; with European point of views again... " We 're taught to draw " what is around or in between "... a flag means you draw the " volume-shape of the wind and the speed of it through the cristal shiny blue of the sky " ...
Isn' t that reality as well ?


Denis Kozlov said...

For 3D artists it's easier - you just set up the forces (gravity and wind), cloth properties and simulate the physical behaviour:) I remember myself doing it for some tv commercial - the results were actually the same as those pencil stketches James did. But then another issue came up - usually (especially in the commercials:) you need the image clear and nice - not just realistic - and this is exactly when those chaotic folds on the free end of the flag become a problem. So after tuning the forces and material properties I ended up simulating a long timespan and picking the best-looking piece for a final animation - just like you'd do with the real footage.